---------------------------------------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Final Log.log
  log type:  text
 opened on:   1 Dec 2023, 15:09:24

. do "/var/folders/c2/tsnfrdmj4q58kvqn2tp5nhw00000gn/T//SD49296.000000"

. /*      *************************************************************/
. /*      File Name:      Final do Models for RR.do                                  */
. /*      Date:           December 1, 2023                          */
. /*      Author:         Kellan Borror                              */   
. /*      Input Files: Final Data.csv                                                     */
. /*      Output File:                                                                */  
. /*      *************************************************************/
. 
. 
. 
. **Set Working Directory and import data**
. cd "/Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1"
/Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1

. import delimited "/Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Final Data.csv"
(encoding automatically selected: ISO-8859-1)
(121 vars, 208 obs)

. 
. **generate logged variables**
. gen log_force_size = ln(force_size)
(19 missing values generated)

. gen log_pop_density = ln(pop_density)
(18 missing values generated)

. gen log_gni_per_cap = ln(gni_per_cap)
(26 missing values generated)

. gen log_gdp_mission = ln(gdp_mission)
(19 missing values generated)

. gen log_pko_density = ln(force_density)
(19 missing values generated)

. 
. 
. **Models with Zero-inflated Binomial Regression**
. **DV= sea_cat1_total, IV= log_ngo_counts**
. 
. **Model 1**
. zinb sea_cat1_total log_ngo_counts pts_mission gender_ratio log_force_size log_pop_density log_gni_per_cap, i
> nflate (log_ngo_counts pts_mission gender_ratio log_force_size log_pop_density log_gni_per_cap) cluster(missi
> on)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -419.27563  
Iteration 2:  Log pseudolikelihood = -409.54704  
Iteration 3:  Log pseudolikelihood = -408.73965  
Iteration 4:  Log pseudolikelihood = -408.71301  
Iteration 5:  Log pseudolikelihood = -408.71293  
Iteration 6:  Log pseudolikelihood = -408.71293  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -408.71293  (not concave)
Iteration 1:  Log pseudolikelihood = -390.95997  
Iteration 2:  Log pseudolikelihood = -361.74367  
Iteration 3:  Log pseudolikelihood = -347.12438  
Iteration 4:  Log pseudolikelihood = -345.00778  
Iteration 5:  Log pseudolikelihood = -344.94993  
Iteration 6:  Log pseudolikelihood = -344.94956  
Iteration 7:  Log pseudolikelihood = -344.94955  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(6)  = 118.78
Log pseudolikelihood = -344.9496                        Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 27 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
 sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
sea_cat1_total  |
 log_ngo_counts |    -.36833   .1474205    -2.50   0.012     -.657269   -.0793911
    pts_mission |   .0574619   .1876215     0.31   0.759    -.3102696    .4251933
   gender_ratio |  -5.078576    4.50021    -1.13   0.259    -13.89883    3.741675
 log_force_size |   .4288427   .0835315     5.13   0.000     .2651239    .5925614
log_pop_density |   .1363278   .0765101     1.78   0.075    -.0136291    .2862848
log_gni_per_cap |  -.6591225   .1302719    -5.06   0.000    -.9144507   -.4037943
          _cons |   4.385876   1.692811     2.59   0.010     1.068028    7.703723
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |   .4239687   .6751336     0.63   0.530    -.8992688    1.747206
    pts_mission |  -.5809327   .8246916    -0.70   0.481    -2.197299    1.035433
   gender_ratio |  -3.197862   19.99518    -0.16   0.873    -42.38769    35.99196
 log_force_size |  -1.027214   .2963761    -3.47   0.001    -1.608101   -.4463279
log_pop_density |   -1.15948   .5723608    -2.03   0.043    -2.281287   -.0376736
log_gni_per_cap |  -.4634157   .4543376    -1.02   0.308    -1.353901    .4270697
          _cons |   12.88599   4.944083     2.61   0.009     3.195768    22.57622
----------------+----------------------------------------------------------------
       /lnalpha |  -1.020606   .2249257    -4.54   0.000    -1.461453   -.5797603
----------------+----------------------------------------------------------------
          alpha |   .3603763   .0810579                      .2318992    .5600326
---------------------------------------------------------------------------------

. 
. **Model 2**
. zinb sea_cat1_total log_ngo_counts corruptionmilvars gender_ratio log_force_size log_pop_density log_gni_per_
> cap pko_fatalities_malicious, inflate (log_ngo_counts corruptionmilvars gender_ratio log_force_size log_pop_d
> ensity log_gni_per_cap pko_fatalities_malicious) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood =  -419.1863  
Iteration 2:  Log pseudolikelihood = -409.00218  
Iteration 3:  Log pseudolikelihood = -407.73281  
Iteration 4:  Log pseudolikelihood = -407.65835  
Iteration 5:  Log pseudolikelihood = -407.65728  
Iteration 6:  Log pseudolikelihood = -407.65728  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -407.65728  (not concave)
Iteration 1:  Log pseudolikelihood =  -385.5628  
Iteration 2:  Log pseudolikelihood = -350.83479  
Iteration 3:  Log pseudolikelihood = -342.80498  
Iteration 4:  Log pseudolikelihood = -341.57651  
Iteration 5:  Log pseudolikelihood = -341.45115  
Iteration 6:  Log pseudolikelihood = -341.44994  
Iteration 7:  Log pseudolikelihood = -341.44994  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(7)  = 154.31
Log pseudolikelihood = -341.4499                        Prob > chi2   = 0.0000

                                           (Std. err. adjusted for 27 clusters in mission)
------------------------------------------------------------------------------------------
                         |               Robust
          sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
sea_cat1_total           |
          log_ngo_counts |  -.3660904   .1479675    -2.47   0.013    -.6561014   -.0760794
       corruptionmilvars |   .3093319    .445596     0.69   0.488    -.5640203    1.182684
            gender_ratio |  -4.000816   4.308533    -0.93   0.353    -12.44539    4.443754
          log_force_size |   .5117185    .141815     3.61   0.000     .2337662    .7896708
         log_pop_density |   .0386583    .121746     0.32   0.751    -.1999595    .2772761
         log_gni_per_cap |  -.6733138   .1221448    -5.51   0.000    -.9127133   -.4339143
pko_fatalities_malicious |   -.015369   .0117854    -1.30   0.192     -.038468    .0077299
                   _cons |   4.347371   1.598603     2.72   0.007     1.214166    7.480575
-------------------------+----------------------------------------------------------------
inflate                  |
          log_ngo_counts |   .7401585   1.058551     0.70   0.484    -1.334564    2.814881
       corruptionmilvars |  -2.045182   3.296617    -0.62   0.535    -8.506433    4.416069
            gender_ratio |  -.1052157   29.84341    -0.00   0.997    -58.59721    58.38678
          log_force_size |  -1.648106   1.266458    -1.30   0.193    -4.130318    .8341064
         log_pop_density |  -1.770243   1.303744    -1.36   0.175    -4.325534    .7850477
         log_gni_per_cap |  -.2936222    .492082    -0.60   0.551    -1.258085    .6708407
pko_fatalities_malicious |  -.2779571   .1465957    -1.90   0.058    -.5652793    .0093652
                   _cons |   15.44139    7.65226     2.02   0.044      .443233    30.43954
-------------------------+----------------------------------------------------------------
                /lnalpha |  -1.038869   .2357901    -4.41   0.000    -1.501009   -.5767289
-------------------------+----------------------------------------------------------------
                   alpha |   .3538546   .0834354                      .2229051    .5617329
------------------------------------------------------------------------------------------

. 
. **Model 3**
. 
. zinb sea_cat1_total log_ngo_counts log_gdp_mission gender_ratio log_force_size log_pop_density log_gni_per_ca
> p, inflate (log_ngo_counts log_gdp_mission gender_ratio log_force_size log_pop_density log_gni_per_cap) diffi
> cult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -419.19226  
Iteration 2:  Log pseudolikelihood = -409.00861  
Iteration 3:  Log pseudolikelihood = -407.82101  
Iteration 4:  Log pseudolikelihood = -407.77609  
Iteration 5:  Log pseudolikelihood = -407.77606  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -407.77606  (not concave)
Iteration 1:  Log pseudolikelihood = -384.18036  
Iteration 2:  Log pseudolikelihood = -346.59506  
Iteration 3:  Log pseudolikelihood = -342.03248  
Iteration 4:  Log pseudolikelihood = -341.39905  
Iteration 5:  Log pseudolikelihood =  -340.8397  
Iteration 6:  Log pseudolikelihood = -340.49415  (not concave)
Iteration 7:  Log pseudolikelihood = -339.18752  
Iteration 8:  Log pseudolikelihood = -338.26461  
Iteration 9:  Log pseudolikelihood = -338.11634  
Iteration 10: Log pseudolikelihood = -338.09173  (not concave)
Iteration 11: Log pseudolikelihood = -338.07741  
Iteration 12: Log pseudolikelihood = -338.04835  
Iteration 13: Log pseudolikelihood = -338.04091  
Iteration 14: Log pseudolikelihood =  -338.0407  
Iteration 15: Log pseudolikelihood =  -338.0407  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(6)  =  63.13
Log pseudolikelihood = -338.0407                        Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 27 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
 sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
sea_cat1_total  |
 log_ngo_counts |  -.5206309   .1407186    -3.70   0.000    -.7964344   -.2448274
log_gdp_mission |  -.0951944   .1519935    -0.63   0.531    -.3930961    .2027073
   gender_ratio |  -6.567262   3.671718    -1.79   0.074     -13.7637    .6291729
 log_force_size |    .559176   .1390279     4.02   0.000     .2866863    .8316657
log_pop_density |   .1644013   .1003552     1.64   0.101    -.0322912    .3610938
log_gni_per_cap |  -.5552444   .1389634    -4.00   0.000    -.8276078   -.2828811
          _cons |   4.077879   2.484088     1.64   0.101    -.7908437    8.946601
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |   -4.49982   10.33717    -0.44   0.663    -24.76031    15.76067
log_gdp_mission |  -7.549855   11.92218    -0.63   0.527    -30.91689    15.81718
   gender_ratio |  -59.50904   84.57517    -0.70   0.482    -225.2733    106.2552
 log_force_size |  -6.568914   7.957236    -0.83   0.409    -22.16481    9.026982
log_pop_density |  -6.868052   7.060268    -0.97   0.331    -20.70592    6.969819
log_gni_per_cap |   .9781372   .5459093     1.79   0.073    -.0918254      2.0481
          _cons |   143.9589   220.1757     0.65   0.513    -287.5776    575.4955
----------------+----------------------------------------------------------------
       /lnalpha |  -.9536331   .2440911    -3.91   0.000    -1.432043   -.4752233
----------------+----------------------------------------------------------------
          alpha |   .3853385   .0940577                      .2388205    .6217462
---------------------------------------------------------------------------------

. 
. **Model 4**
. 
. zinb sea_cat1_total log_ngo_counts pts_mission gender_ratio log_pop_density force_density, inflate (log_ngo_c
> ounts pts_mission gender_ratio log_pop_density force_density) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -518.48954  
Iteration 1:  Log pseudolikelihood = -467.93399  
Iteration 2:  Log pseudolikelihood = -462.81191  
Iteration 3:  Log pseudolikelihood = -458.12336  
Iteration 4:  Log pseudolikelihood = -456.06293  
Iteration 5:  Log pseudolikelihood = -453.00897  
Iteration 6:  Log pseudolikelihood = -447.26253  
Iteration 7:  Log pseudolikelihood = -444.25347  
Iteration 8:  Log pseudolikelihood = -444.15111  
Iteration 9:  Log pseudolikelihood = -444.15077  
Iteration 10: Log pseudolikelihood = -444.15077  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -444.15077  (not concave)
Iteration 1:  Log pseudolikelihood = -435.29919  
Iteration 2:  Log pseudolikelihood = -410.86493  
Iteration 3:  Log pseudolikelihood = -407.71322  
Iteration 4:  Log pseudolikelihood = -407.68844  
Iteration 5:  Log pseudolikelihood = -407.68843  

Zero-inflated negative binomial regression              Number of obs =    188
Inflation model: logit                                  Nonzero obs   =    108
                                                        Zero obs      =     80
                                                        Wald chi2(5)  =  33.18
Log pseudolikelihood = -407.6884                        Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 30 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
 sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
sea_cat1_total  |
 log_ngo_counts |   .1240777   .1544157     0.80   0.422    -.1785714    .4267268
    pts_mission |   1.536292   .3196024     4.81   0.000     .9098833    2.162701
   gender_ratio |  -19.86255   7.657052    -2.59   0.009     -34.8701   -4.855003
log_pop_density |   .0284791    .165284     0.17   0.863    -.2954715    .3524297
  force_density |   .1714264   .7373476     0.23   0.816    -1.273748    1.616601
          _cons |  -1.368298    .924976    -1.48   0.139    -3.181218    .4446213
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |  -.3148405   .4810387    -0.65   0.513    -1.257659    .6279781
    pts_mission |  -.1244728   .4920152    -0.25   0.800    -1.088805    .8398592
   gender_ratio |  -7.749105   8.082489    -0.96   0.338    -23.59049    8.092282
log_pop_density |   .5539056    .428564     1.29   0.196    -.2860643    1.393876
  force_density |  -496.9183   121.2119    -4.10   0.000    -734.4893   -259.3474
          _cons |   1.329848   1.442619     0.92   0.357    -1.497633    4.157328
----------------+----------------------------------------------------------------
       /lnalpha |  -.1581482   .1760421    -0.90   0.369    -.5031844    .1868881
----------------+----------------------------------------------------------------
          alpha |   .8537233   .1502913                      .6046023    1.205492
---------------------------------------------------------------------------------

. 
. **Model 5** 
. 
. zinb sea_cat1_total log_ngo_counts corruptionmilvars force_qual_mil gender_ratio log_pop_density force_densit
> y, inflate (log_ngo_counts corruptionmilvars force_qual_mil gender_ratio log_pop_density force_density) diffi
> cult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -518.48954  
Iteration 1:  Log pseudolikelihood = -473.83967  
Iteration 2:  Log pseudolikelihood = -468.13269  
Iteration 3:  Log pseudolikelihood = -463.05967  
Iteration 4:  Log pseudolikelihood = -457.83246  
Iteration 5:  Log pseudolikelihood = -451.79982  
Iteration 6:  Log pseudolikelihood =  -446.3118  
Iteration 7:  Log pseudolikelihood = -444.50343  
Iteration 8:  Log pseudolikelihood = -444.48016  
Iteration 9:  Log pseudolikelihood = -444.48014  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -444.48014  (not concave)
Iteration 1:  Log pseudolikelihood = -433.05632  
Iteration 2:  Log pseudolikelihood = -419.83502  
Iteration 3:  Log pseudolikelihood = -416.26076  
Iteration 4:  Log pseudolikelihood = -406.89168  
Iteration 5:  Log pseudolikelihood = -406.76443  
Iteration 6:  Log pseudolikelihood = -406.76423  
Iteration 7:  Log pseudolikelihood = -406.76423  

Zero-inflated negative binomial regression              Number of obs =    188
Inflation model: logit                                  Nonzero obs   =    108
                                                        Zero obs      =     80
                                                        Wald chi2(6)  =  45.92
Log pseudolikelihood = -406.7642                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 30 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
   log_ngo_counts |  -.1763333   .1765128    -1.00   0.318     -.522292    .1696253
corruptionmilvars |  -2.140926   .4491244    -4.77   0.000    -3.021194   -1.260659
   force_qual_mil |  -.0122309   .3310382    -0.04   0.971    -.6610539    .6365922
     gender_ratio |   -22.2725   6.539541    -3.41   0.001    -35.08976   -9.455234
  log_pop_density |   .4802946   .2412585     1.99   0.047     .0074367    .9531525
    force_density |  -.6889924    .753437    -0.91   0.360    -2.165702    .7877171
            _cons |   1.656347   .5117043     3.24   0.001      .653425    2.659269
------------------+----------------------------------------------------------------
inflate           |
   log_ngo_counts |  -.5749116   .8697319    -0.66   0.509    -2.279555    1.129732
corruptionmilvars |  -1.628279   1.124281    -1.45   0.148    -3.831829    .5752708
   force_qual_mil |   .0761948   .4272562     0.18   0.858     -.761212    .9136016
     gender_ratio |  -12.37769   8.254256    -1.50   0.134    -28.55573    3.800356
  log_pop_density |   1.061562   .7276931     1.46   0.145    -.3646901    2.487815
    force_density |  -642.7305   305.8026    -2.10   0.036    -1242.093    -43.3684
            _cons |   .8532996   1.203876     0.71   0.478    -1.506254    3.212853
------------------+----------------------------------------------------------------
         /lnalpha |  -.0835889   .1792328    -0.47   0.641    -.4348787    .2677009
------------------+----------------------------------------------------------------
            alpha |   .9198093     .16486                      .6473432    1.306956
-----------------------------------------------------------------------------------

. 
. **Model 6**
. 
. zinb sea_cat1_total log_ngo_counts pts_mission force_qual_mil gender_ratio log_force_size ln_km2_country , in
> flate (log_ngo_counts pts_mission force_qual_mil gender_ratio log_force_size ln_km2_country) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -518.48954  
Iteration 1:  Log pseudolikelihood =   -455.598  
Iteration 2:  Log pseudolikelihood = -445.19491  
Iteration 3:  Log pseudolikelihood = -444.81078  
Iteration 4:  Log pseudolikelihood = -444.80639  
Iteration 5:  Log pseudolikelihood = -444.80638  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -444.80638  (not concave)
Iteration 1:  Log pseudolikelihood = -429.72936  
Iteration 2:  Log pseudolikelihood = -399.89011  
Iteration 3:  Log pseudolikelihood =  -389.6834  
Iteration 4:  Log pseudolikelihood = -389.62758  
Iteration 5:  Log pseudolikelihood = -385.58785  
Iteration 6:  Log pseudolikelihood = -384.41101  
Iteration 7:  Log pseudolikelihood = -383.49004  
Iteration 8:  Log pseudolikelihood = -383.32187  
Iteration 9:  Log pseudolikelihood = -383.24872  
Iteration 10: Log pseudolikelihood = -383.24595  
Iteration 11: Log pseudolikelihood = -383.24594  

Zero-inflated negative binomial regression              Number of obs =    188
Inflation model: logit                                  Nonzero obs   =    108
                                                        Zero obs      =     80
                                                        Wald chi2(6)  = 120.81
Log pseudolikelihood = -383.2459                        Prob > chi2   = 0.0000

                                 (Std. err. adjusted for 30 clusters in mission)
--------------------------------------------------------------------------------
               |               Robust
sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
---------------+----------------------------------------------------------------
sea_cat1_total |
log_ngo_counts |  -.4178465   .1523541    -2.74   0.006    -.7164549    -.119238
   pts_mission |   .7925628   .2995474     2.65   0.008     .2054606    1.379665
force_qual_mil |  -.2346489   .2510601    -0.93   0.350    -.7267177      .25742
  gender_ratio |  -14.43683    6.79647    -2.12   0.034    -27.75767   -1.115997
log_force_size |   .6537286   .1704158     3.84   0.000     .3197198    .9877375
ln_km2_country |   .0033577   .1070603     0.03   0.975    -.2064767     .213192
         _cons |  -2.881088   1.267138    -2.27   0.023    -5.364632    -.397543
---------------+----------------------------------------------------------------
inflate        |
log_ngo_counts |  -2.467889   .6547487    -3.77   0.000    -3.751172   -1.184605
   pts_mission |   2.351904   2.361791     1.00   0.319     -2.27712    6.980929
force_qual_mil |   11.45551   2.592907     4.42   0.000     6.373508    16.53752
  gender_ratio |  -326.4284   97.70476    -3.34   0.001    -517.9262   -134.9306
log_force_size |  -7.735982   2.216982    -3.49   0.000    -12.08119   -3.390777
ln_km2_country |   8.118286   2.295591     3.54   0.000     3.619011    12.61756
         _cons |  -51.58213   17.04339    -3.03   0.002    -84.98657   -18.17769
---------------+----------------------------------------------------------------
      /lnalpha |  -.4239804   .2601242    -1.63   0.103    -.9338145    .0858538
---------------+----------------------------------------------------------------
         alpha |   .6544367   .1702349                      .3930516    1.089647
--------------------------------------------------------------------------------

. 
. 
. **Model 7**
. 
. zinb sea_cat1_total log_ngo_counts corruptionmilvars gender_ratio log_force_size ln_km2_country, inflate (log
> _ngo_counts corruptionmilvars gender_ratio log_force_size ln_km2_country) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -518.48954  
Iteration 1:  Log pseudolikelihood = -456.80576  
Iteration 2:  Log pseudolikelihood = -448.07813  
Iteration 3:  Log pseudolikelihood = -447.78939  
Iteration 4:  Log pseudolikelihood = -447.78858  
Iteration 5:  Log pseudolikelihood = -447.78858  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -447.78858  (not concave)
Iteration 1:  Log pseudolikelihood = -431.89974  
Iteration 2:  Log pseudolikelihood = -417.23998  
Iteration 3:  Log pseudolikelihood = -399.17307  
Iteration 4:  Log pseudolikelihood = -396.40122  
Iteration 5:  Log pseudolikelihood = -394.13494  
Iteration 6:  Log pseudolikelihood = -393.63474  
Iteration 7:  Log pseudolikelihood = -393.55453  
Iteration 8:  Log pseudolikelihood = -393.55132  
Iteration 9:  Log pseudolikelihood =  -393.5513  

Zero-inflated negative binomial regression              Number of obs =    188
Inflation model: logit                                  Nonzero obs   =    108
                                                        Zero obs      =     80
                                                        Wald chi2(5)  =  89.15
Log pseudolikelihood = -393.5513                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 30 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
   log_ngo_counts |  -.4738205   .1892436    -2.50   0.012    -.8447311   -.1029098
corruptionmilvars |  -.1906694    .422076    -0.45   0.651    -1.017923    .6365843
     gender_ratio |  -12.43393   7.386147    -1.68   0.092    -26.91051    2.042654
   log_force_size |   .7514966   .2307282     3.26   0.001     .2992777    1.203716
   ln_km2_country |  -.0246852   .1334005    -0.19   0.853    -.2861453    .2367749
            _cons |  -1.414932   1.815268    -0.78   0.436    -4.972791    2.142928
------------------+----------------------------------------------------------------
inflate           |
   log_ngo_counts |   .4192194   .8440065     0.50   0.619    -1.235003    2.073442
corruptionmilvars |  -9.659089   7.398669    -1.31   0.192    -24.16021    4.842036
     gender_ratio |  -45.66452   46.90638    -0.97   0.330    -137.5993     46.2703
   log_force_size |  -3.462119   2.346806    -1.48   0.140    -8.061774    1.137535
   ln_km2_country |   2.924341   2.432777     1.20   0.229    -1.843815    7.692497
            _cons |  -19.89281   24.67377    -0.81   0.420    -68.25251     28.4669
------------------+----------------------------------------------------------------
         /lnalpha |  -.3124422   .3124529    -1.00   0.317    -.9248386    .2999543
------------------+----------------------------------------------------------------
            alpha |   .7316579   .2286086                      .3965954    1.349797
-----------------------------------------------------------------------------------

. 
. 
. **Models with Zero-inflated Binomial Regression**
. **DV= sea_cat1_total, IV= log_ngo_counts_hr**
. 
. **Model 8**
. zinb sea_cat1_total log_ngo_counts_hr pts_mission gender_ratio log_force_size log_gni_per_cap, inflate (log_n
> go_counts_hr pts_mission gender_ratio log_force_size log_gni_per_cap) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -419.65931  
Iteration 2:  Log pseudolikelihood = -411.02426  
Iteration 3:  Log pseudolikelihood = -410.64733  
Iteration 4:  Log pseudolikelihood = -410.64209  
Iteration 5:  Log pseudolikelihood = -410.64208  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -410.64208  (not concave)
Iteration 1:  Log pseudolikelihood = -392.81625  
Iteration 2:  Log pseudolikelihood = -357.44927  (backed up)
Iteration 3:  Log pseudolikelihood = -350.37621  
Iteration 4:  Log pseudolikelihood = -349.95929  
Iteration 5:  Log pseudolikelihood = -349.95573  
Iteration 6:  Log pseudolikelihood = -349.95573  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(5)  = 153.06
Log pseudolikelihood = -349.9557                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 27 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
log_ngo_counts_hr |  -.4241099   .2004736    -2.12   0.034    -.8170308   -.0311889
      pts_mission |  -.0261804   .2010457    -0.13   0.896    -.4202228    .3678619
     gender_ratio |   -4.18209   4.209002    -0.99   0.320    -12.43158    4.067403
   log_force_size |   .4345677   .0828233     5.25   0.000      .272237    .5968983
  log_gni_per_cap |  -.5981868   .1166087    -5.13   0.000    -.8267356    -.369638
            _cons |   4.453133   1.555192     2.86   0.004     1.405013    7.501253
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |    .165482   .5491179     0.30   0.763    -.9107692    1.241733
      pts_mission |  -1.100595   .8386933    -1.31   0.189    -2.744404    .5432134
     gender_ratio |  -3.677691   13.14812    -0.28   0.780    -29.44753    22.09215
   log_force_size |   -.635038   .2166404    -2.93   0.003    -1.059645   -.2104305
  log_gni_per_cap |  -.6728958   .3678973    -1.83   0.067    -1.393961    .0481697
            _cons |   9.588462   5.184891     1.85   0.064    -.5737382    19.75066
------------------+----------------------------------------------------------------
         /lnalpha |  -.9153787   .2015872    -4.54   0.000    -1.310482   -.5202751
------------------+----------------------------------------------------------------
            alpha |    .400365   .0807084                      .2696899     .594357
-----------------------------------------------------------------------------------

. 
. **Model 9**
. zinb sea_cat1_total log_ngo_counts_hr corruptionmilvars gender_ratio log_force_size log_pop_density log_gni_p
> er_cap pko_fatalities_malicious, inflate (log_ngo_counts_hr corruptionmilvars gender_ratio log_force_size log
> _pop_density log_gni_per_cap pko_fatalities_malicious) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -418.93187  
Iteration 2:  Log pseudolikelihood = -408.57655  
Iteration 3:  Log pseudolikelihood =  -407.1834  
Iteration 4:  Log pseudolikelihood = -407.10282  
Iteration 5:  Log pseudolikelihood = -407.10197  
Iteration 6:  Log pseudolikelihood = -407.10197  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -407.10197  (not concave)
Iteration 1:  Log pseudolikelihood = -385.03702  
Iteration 2:  Log pseudolikelihood = -350.49999  
Iteration 3:  Log pseudolikelihood = -342.35584  
Iteration 4:  Log pseudolikelihood = -341.07191  
Iteration 5:  Log pseudolikelihood = -340.93469  
Iteration 6:  Log pseudolikelihood = -340.93332  
Iteration 7:  Log pseudolikelihood = -340.93332  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(7)  = 172.42
Log pseudolikelihood = -340.9333                        Prob > chi2   = 0.0000

                                           (Std. err. adjusted for 27 clusters in mission)
------------------------------------------------------------------------------------------
                         |               Robust
          sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
sea_cat1_total           |
       log_ngo_counts_hr |  -.4341613    .210828    -2.06   0.039    -.8473765    -.020946
       corruptionmilvars |   .2840192    .441775     0.64   0.520    -.5818438    1.149882
            gender_ratio |  -4.181641   4.223622    -0.99   0.322    -12.45979    4.096507
          log_force_size |   .5172346   .1433735     3.61   0.000     .2362278    .7982414
         log_pop_density |   .0339576   .1190526     0.29   0.775    -.1993813    .2672965
         log_gni_per_cap |  -.6486384   .1205896    -5.38   0.000    -.8849897   -.4122871
pko_fatalities_malicious |  -.0176748   .0115434    -1.53   0.126    -.0402995    .0049499
                   _cons |   4.023714    1.62873     2.47   0.013     .8314614    7.215967
-------------------------+----------------------------------------------------------------
inflate                  |
       log_ngo_counts_hr |   1.156508   1.111762     1.04   0.298    -1.022506    3.335522
       corruptionmilvars |  -2.023496   3.233414    -0.63   0.531    -8.360872    4.313879
            gender_ratio |  -.9518081   30.97728    -0.03   0.975    -61.66615    59.76254
          log_force_size |  -1.687563   1.249873    -1.35   0.177    -4.137269    .7621436
         log_pop_density |  -1.897525   1.215284    -1.56   0.118    -4.279438    .4843882
         log_gni_per_cap |  -.2918286   .5315617    -0.55   0.583     -1.33367    .7500132
pko_fatalities_malicious |  -.2751009   .1459357    -1.89   0.059    -.5611296    .0109278
                   _cons |   15.39988   7.533923     2.04   0.041     .6336605    30.16609
-------------------------+----------------------------------------------------------------
                /lnalpha |  -1.028504   .2326096    -4.42   0.000     -1.48441   -.5725973
-------------------------+----------------------------------------------------------------
                   alpha |   .3575416   .0831676                       .226636    .5640585
------------------------------------------------------------------------------------------

. 
. **Model 10**
. 
. zinb sea_cat1_total log_ngo_counts_hr log_gdp_mission gender_ratio log_force_size log_pop_density ln_km2_coun
> try, inflate (log_ngo_counts_hr log_gdp_mission gender_ratio log_force_size log_pop_density ln_km2_country) d
> ifficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -518.48954  
Iteration 1:  Log pseudolikelihood = -456.79258  
Iteration 2:  Log pseudolikelihood = -447.76347  
Iteration 3:  Log pseudolikelihood = -447.41853  
Iteration 4:  Log pseudolikelihood = -447.41689  
Iteration 5:  Log pseudolikelihood = -447.41689  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -447.41689  (not concave)
Iteration 1:  Log pseudolikelihood = -431.10025  
Iteration 2:  Log pseudolikelihood = -407.09236  
Iteration 3:  Log pseudolikelihood = -391.63738  
Iteration 4:  Log pseudolikelihood = -387.10792  
Iteration 5:  Log pseudolikelihood = -384.47351  
Iteration 6:  Log pseudolikelihood = -383.32838  
Iteration 7:  Log pseudolikelihood = -383.08751  
Iteration 8:  Log pseudolikelihood = -383.06241  
Iteration 9:  Log pseudolikelihood =  -383.0612  
Iteration 10: Log pseudolikelihood =  -383.0612  

Zero-inflated negative binomial regression              Number of obs =    188
Inflation model: logit                                  Nonzero obs   =    108
                                                        Zero obs      =     80
                                                        Wald chi2(6)  = 121.72
Log pseudolikelihood = -383.0612                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 30 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
log_ngo_counts_hr |  -.8008332    .286601    -2.79   0.005    -1.362561   -.2391055
  log_gdp_mission |  -.4606559   .2517314    -1.83   0.067    -.9540405    .0327286
     gender_ratio |   -11.8966   7.107148    -1.67   0.094    -25.82636    2.033153
   log_force_size |   .6596026   .1965644     3.36   0.001     .2743435    1.044862
  log_pop_density |   .2323431   .2004948     1.16   0.247    -.1606195    .6253057
   ln_km2_country |   .0329404   .1522111     0.22   0.829    -.2653879    .3312688
            _cons |   2.401462   3.528333     0.68   0.496    -4.513943    9.316868
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |  -6.712619   1.389412    -4.83   0.000    -9.435817   -3.989421
  log_gdp_mission |  -8.105208   2.340069    -3.46   0.001    -12.69166   -3.518756
     gender_ratio |  -108.8907   30.38871    -3.58   0.000    -168.4515   -49.32991
   log_force_size |  -5.149364   1.768221    -2.91   0.004    -8.615013   -1.683715
  log_pop_density |   2.704831   .7941484     3.41   0.001     1.148329    4.261333
   ln_km2_country |   5.414467   1.192799     4.54   0.000     3.076624     7.75231
            _cons |   43.80436   20.24142     2.16   0.030       4.1319    83.47681
------------------+----------------------------------------------------------------
         /lnalpha |  -.4487135   .2711967    -1.65   0.098    -.9802493    .0828224
------------------+----------------------------------------------------------------
            alpha |    .638449   .1731453                      .3752176    1.086349
-----------------------------------------------------------------------------------

. 
. **Model 11**
. 
. zinb sea_cat1_total log_ngo_counts_hr pts_mission force_qual_mil gender_ratio log_force_size log_gni_per_cap,
>  inflate (log_ngo_counts_hr pts_mission force_qual_mil gender_ratio log_force_size log_gni_per_cap ) difficul
> t cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -417.67479  
Iteration 2:  Log pseudolikelihood = -407.19469  
Iteration 3:  Log pseudolikelihood = -406.63652  
Iteration 4:  Log pseudolikelihood = -406.62696  
Iteration 5:  Log pseudolikelihood = -406.62696  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -406.62696  (not concave)
Iteration 1:  Log pseudolikelihood = -383.60641  
Iteration 2:  Log pseudolikelihood = -347.13996  
Iteration 3:  Log pseudolikelihood =  -339.9886  
Iteration 4:  Log pseudolikelihood = -339.18097  
Iteration 5:  Log pseudolikelihood = -339.12178  
Iteration 6:  Log pseudolikelihood = -339.12148  
Iteration 7:  Log pseudolikelihood = -339.12148  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(6)  = 188.11
Log pseudolikelihood = -339.1215                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 27 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
log_ngo_counts_hr |   -.471257   .2001176    -2.35   0.019    -.8634802   -.0790337
      pts_mission |   .1606429   .1710437     0.94   0.348    -.1745966    .4958824
   force_qual_mil |  -.2486537   .1431748    -1.74   0.082    -.5292712    .0319638
     gender_ratio |  -5.125566    3.24463    -1.58   0.114    -11.48493    1.233793
   log_force_size |    .485022   .0899629     5.39   0.000      .308698     .661346
  log_gni_per_cap |  -.5898634    .121724    -4.85   0.000     -.828438   -.3512887
            _cons |   3.789386   1.673622     2.26   0.024     .5091458    7.069625
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |   .3945638   .7013671     0.56   0.574    -.9800904    1.769218
      pts_mission |  -1.747492   .9570546    -1.83   0.068    -3.623284    .1283008
   force_qual_mil |   2.117866   .9855487     2.15   0.032     .1862259    4.049506
     gender_ratio |  -1.913895   13.97847    -0.14   0.891     -29.3112    25.48341
   log_force_size |  -1.132726   .4776462    -2.37   0.018    -2.068895   -.1965561
  log_gni_per_cap |  -1.237119   .8283098    -1.49   0.135    -2.860576    .3863389
            _cons |    16.4288   7.973937     2.06   0.039     .8001728    32.05743
------------------+----------------------------------------------------------------
         /lnalpha |   -1.01088   .2076892    -4.87   0.000    -1.417943   -.6038163
------------------+----------------------------------------------------------------
            alpha |   .3638987   .0755778                      .2422117    .5467212
-----------------------------------------------------------------------------------

. 
. 
. 
. **Models with Zero-inflated Binomial Regression**
. **DV= sea_total, IV= log_ngo_counts_w**
. 
. **Model 12**
. zinb sea_total log_ngo_counts_w pts_mission  gender_ratio log_force_size log_pop_density log_gni_per_cap , in
> flate (log_ngo_counts_w pts_mission gender_ratio log_force_size log_pop_density log_gni_per_cap  ) difficult 
> cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -338.20938  (not concave)
Iteration 1:  Log pseudolikelihood = -271.33666  
Iteration 2:  Log pseudolikelihood = -269.45494  
Iteration 3:  Log pseudolikelihood =  -269.3951  
Iteration 4:  Log pseudolikelihood = -269.39493  
Iteration 5:  Log pseudolikelihood = -269.39493  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -269.39493  (not concave)
Iteration 1:  Log pseudolikelihood = -246.52504  (not concave)
Iteration 2:  Log pseudolikelihood = -230.95256  (not concave)
Iteration 3:  Log pseudolikelihood =   -224.612  
Iteration 4:  Log pseudolikelihood = -220.78869  
Iteration 5:  Log pseudolikelihood = -220.07508  
Iteration 6:  Log pseudolikelihood = -219.78068  
Iteration 7:  Log pseudolikelihood = -219.68811  
Iteration 8:  Log pseudolikelihood = -219.68705  
Iteration 9:  Log pseudolikelihood = -219.68704  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     69
                                                        Zero obs      =     96
                                                        Wald chi2(6)  =  77.98
Log pseudolikelihood = -219.687                         Prob > chi2   = 0.0000

                                   (Std. err. adjusted for 27 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
       sea_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_total        |
log_ngo_counts_w |  -.2746243   .2240927    -1.23   0.220     -.713838    .1645893
     pts_mission |    .291087   .2966418     0.98   0.326    -.2903203    .8724943
    gender_ratio |  -4.294218   10.52344    -0.41   0.683    -24.91978    16.33134
  log_force_size |   .3128914   .2282017     1.37   0.170    -.1343757    .7601586
 log_pop_density |   .1786824   .1599524     1.12   0.264    -.1348186    .4921834
 log_gni_per_cap |  -1.234051   .2633893    -4.69   0.000    -1.750284   -.7178172
           _cons |   6.578112   2.763101     2.38   0.017     1.162534    11.99369
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |   2.281688   4.967398     0.46   0.646    -7.454234    12.01761
     pts_mission |   1.574281   1.467459     1.07   0.283    -1.301886    4.450448
    gender_ratio |   42.78145   69.65718     0.61   0.539    -93.74411     179.307
  log_force_size |  -2.469222   1.043094    -2.37   0.018    -4.513649   -.4247953
 log_pop_density |  -3.418284    2.30704    -1.48   0.138        -7.94    1.103431
 log_gni_per_cap |  -1.637711   1.391216    -1.18   0.239    -4.364444    1.089022
           _cons |   28.34446   19.25637     1.47   0.141    -9.397339    66.08625
-----------------+----------------------------------------------------------------
        /lnalpha |  -1.141814   .3098731    -3.68   0.000    -1.749154   -.5344742
-----------------+----------------------------------------------------------------
           alpha |   .3192393   .0989237                       .173921    .5859773
----------------------------------------------------------------------------------

. 
. **Model 13**
. zinb sea_total log_ngo_counts_w corruptionmilvars gender_ratio log_force_size log_pop_density pko_fatalities_
> malicious log_gni_per_cap, inflate (log_ngo_counts_w corruptionmilvars gender_ratio log_force_size log_pop_de
> nsity pko_fatalities_malicious log_gni_per_cap) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -338.20938  (not concave)
Iteration 1:  Log pseudolikelihood = -271.23416  
Iteration 2:  Log pseudolikelihood =  -269.3632  
Iteration 3:  Log pseudolikelihood = -269.31958  
Iteration 4:  Log pseudolikelihood = -269.31944  
Iteration 5:  Log pseudolikelihood = -269.31944  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -269.31944  (not concave)
Iteration 1:  Log pseudolikelihood =  -246.9533  (not concave)
Iteration 2:  Log pseudolikelihood = -227.51429  
Iteration 3:  Log pseudolikelihood = -220.69015  
Iteration 4:  Log pseudolikelihood = -218.45693  
Iteration 5:  Log pseudolikelihood = -217.47567  
Iteration 6:  Log pseudolikelihood = -217.18934  
Iteration 7:  Log pseudolikelihood = -217.15645  
Iteration 8:  Log pseudolikelihood =  -217.1553  
Iteration 9:  Log pseudolikelihood =  -217.1553  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     69
                                                        Zero obs      =     96
                                                        Wald chi2(7)  =  68.36
Log pseudolikelihood = -217.1553                        Prob > chi2   = 0.0000

                                           (Std. err. adjusted for 27 clusters in mission)
------------------------------------------------------------------------------------------
                         |               Robust
               sea_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
sea_total                |
        log_ngo_counts_w |  -.3155995   .3226608    -0.98   0.328    -.9480032    .3168041
       corruptionmilvars |  -.1497603   .7110613    -0.21   0.833    -1.543415    1.243894
            gender_ratio |  -3.565845   9.841794    -0.36   0.717    -22.85541    15.72372
          log_force_size |    .372419    .487218     0.76   0.445    -.5825107    1.327349
         log_pop_density |   .1393153   .2039895     0.68   0.495    -.2604968    .5391273
pko_fatalities_malicious |  -.0309002   .0217975    -1.42   0.156    -.0736224    .0118221
         log_gni_per_cap |  -1.213132   .2589086    -4.69   0.000    -1.720584   -.7056806
                   _cons |   6.892874   4.377265     1.57   0.115    -1.686407    15.47216
-------------------------+----------------------------------------------------------------
inflate                  |
        log_ngo_counts_w |   .8464015   5.218053     0.16   0.871    -9.380794     11.0736
       corruptionmilvars |  -5.217764   4.310458    -1.21   0.226    -13.66611    3.230578
            gender_ratio |   44.48534   98.45938     0.45   0.651    -148.4915    237.4622
          log_force_size |  -3.867396   2.214258    -1.75   0.081    -8.207263    .4724705
         log_pop_density |  -4.398522   4.680763    -0.94   0.347    -13.57265    4.775605
pko_fatalities_malicious |  -.5875107   1.314122    -0.45   0.655    -3.163143    1.988122
         log_gni_per_cap |  -1.186636   4.689038    -0.25   0.800    -10.37698     8.00371
                   _cons |   46.46911   42.01555     1.11   0.269    -35.87986    128.8181
-------------------------+----------------------------------------------------------------
                /lnalpha |  -1.151097   .3486465    -3.30   0.001    -1.834432   -.4677626
-------------------------+----------------------------------------------------------------
                   alpha |   .3162896   .1102732                      .1597042    .6264022
------------------------------------------------------------------------------------------

. 
. **Model 14**
. 
. zinb sea_total log_ngo_counts_w log_gdp_mission gender_ratio log_force_size log_pop_density , inflate (log_ng
> o_counts_w log_gdp_mission gender_ratio log_force_size log_pop_density ) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -361.04504  (not concave)
Iteration 1:  Log pseudolikelihood = -289.35072  
Iteration 2:  Log pseudolikelihood =  -287.6034  
Iteration 3:  Log pseudolikelihood = -285.49861  
Iteration 4:  Log pseudolikelihood = -285.21795  
Iteration 5:  Log pseudolikelihood = -285.21588  
Iteration 6:  Log pseudolikelihood = -285.21587  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -285.21587  (not concave)
Iteration 1:  Log pseudolikelihood =  -277.5527  (not concave)
Iteration 2:  Log pseudolikelihood =  -262.1225  
Iteration 3:  Log pseudolikelihood = -253.20413  
Iteration 4:  Log pseudolikelihood = -247.73255  
Iteration 5:  Log pseudolikelihood = -243.91003  
Iteration 6:  Log pseudolikelihood = -242.33321  
Iteration 7:  Log pseudolikelihood = -241.36927  
Iteration 8:  Log pseudolikelihood = -240.79248  
Iteration 9:  Log pseudolikelihood = -240.58901  
Iteration 10: Log pseudolikelihood = -240.52011  
Iteration 11: Log pseudolikelihood = -240.51831  
Iteration 12: Log pseudolikelihood =  -240.5183  

Zero-inflated negative binomial regression                 Number of obs = 188
Inflation model: logit                                     Nonzero obs   =  71
                                                           Zero obs      = 117
                                                           Wald chi2(5)  =   .
Log pseudolikelihood = -240.5183                           Prob > chi2   =   .

                                   (Std. err. adjusted for 30 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
       sea_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_total        |
log_ngo_counts_w |  -.6656538   .2147059    -3.10   0.002     -1.08647    -.244838
 log_gdp_mission |   .0728377   .2631018     0.28   0.782    -.4428324    .5885078
    gender_ratio |  -2.622926   5.863469    -0.45   0.655    -14.11511    8.869261
  log_force_size |   1.503166   .2736197     5.49   0.000     .9668816    2.039451
 log_pop_density |   .0663112   .1331025     0.50   0.618    -.1945649    .3271872
           _cons |  -10.79813   3.518239    -3.07   0.002    -17.69375   -3.902506
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |   65.24335   13.42601     4.86   0.000     38.92885    91.55785
 log_gdp_mission |   5.336355   2.863645     1.86   0.062    -.2762872      10.949
    gender_ratio |   26.91664   134.0121     0.20   0.841    -235.7423    289.5756
  log_force_size |    12.4951   4.054778     3.08   0.002     4.547884    20.44232
 log_pop_density |   .9184264   1.156652     0.79   0.427     -1.34857    3.185423
           _cons |  -383.8776   57.93554    -6.63   0.000    -497.4292    -270.326
-----------------+----------------------------------------------------------------
        /lnalpha |   -.498879   .4447469    -1.12   0.262    -1.370567    .3728089
-----------------+----------------------------------------------------------------
           alpha |    .607211   .2700552                      .2539629    1.451807
----------------------------------------------------------------------------------

. 
. **Model 15**
. 
. zinb sea_total log_ngo_counts_w pts_mission force_qual_mil gender_ratio log_force_size ln_km2_country log_gni
> _per_cap , inflate (log_ngo_counts_w pts_mission force_qual_mil gender_ratio log_force_size ln_km2_country lo
> g_gni_per_cap ) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -338.20938  (not concave)
Iteration 1:  Log pseudolikelihood = -293.93225  (not concave)
Iteration 2:  Log pseudolikelihood = -288.15552  
Iteration 3:  Log pseudolikelihood = -272.72732  
Iteration 4:  Log pseudolikelihood =   -264.564  
Iteration 5:  Log pseudolikelihood = -264.28481  
Iteration 6:  Log pseudolikelihood = -264.27102  
Iteration 7:  Log pseudolikelihood = -264.27069  
Iteration 8:  Log pseudolikelihood = -264.27068  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -264.27068  (not concave)
Iteration 1:  Log pseudolikelihood = -256.46016  (not concave)
Iteration 2:  Log pseudolikelihood = -251.49807  (not concave)
Iteration 3:  Log pseudolikelihood =  -248.1355  (not concave)
Iteration 4:  Log pseudolikelihood =  -246.4216  (not concave)
Iteration 5:  Log pseudolikelihood = -241.24643  
Iteration 6:  Log pseudolikelihood = -225.89103  
Iteration 7:  Log pseudolikelihood = -212.45999  
Iteration 8:  Log pseudolikelihood = -210.63095  
Iteration 9:  Log pseudolikelihood = -209.97653  
Iteration 10: Log pseudolikelihood = -208.56717  
Iteration 11: Log pseudolikelihood = -206.77127  
Iteration 12: Log pseudolikelihood = -206.45308  
Iteration 13: Log pseudolikelihood = -206.44299  
Iteration 14: Log pseudolikelihood = -206.44291  
Iteration 15: Log pseudolikelihood = -206.44291  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     69
                                                        Zero obs      =     96
                                                        Wald chi2(7)  =  75.41
Log pseudolikelihood = -206.4429                        Prob > chi2   = 0.0000

                                   (Std. err. adjusted for 27 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
       sea_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_total        |
log_ngo_counts_w |   -.180996   .1861454    -0.97   0.331    -.5458343    .1838424
     pts_mission |   .0782252   .1972851     0.40   0.692    -.3084464    .4648969
  force_qual_mil |   .0460775   .3259255     0.14   0.888    -.5927248    .6848798
    gender_ratio |   2.857076   6.783657     0.42   0.674    -10.43865     16.1528
  log_force_size |   .5073811   .1642702     3.09   0.002     .1854174    .8293449
  ln_km2_country |  -.0395998   .0837387    -0.47   0.636    -.2037247    .1245251
 log_gni_per_cap |  -1.005759    .214642    -4.69   0.000     -1.42645   -.5850688
           _cons |   4.545978    2.75425     1.65   0.099    -.8522535    9.944209
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |   -7.00369   1.644464    -4.26   0.000    -10.22678   -3.780599
     pts_mission |   -4.40962   .6821064    -6.46   0.000    -5.746524   -3.072716
  force_qual_mil |   30.93932   5.802902     5.33   0.000     19.56585     42.3128
    gender_ratio |   148.5703   33.55624     4.43   0.000     82.80127    214.3393
  log_force_size |  -13.68178   2.659973    -5.14   0.000    -18.89523   -8.468328
  ln_km2_country |   7.920026   1.608969     4.92   0.000     4.766505    11.07355
 log_gni_per_cap |   6.370633   1.267506     5.03   0.000     3.886366      8.8549
           _cons |  -38.31129    8.57777    -4.47   0.000    -55.12341   -21.49917
-----------------+----------------------------------------------------------------
        /lnalpha |  -1.153193   .2902979    -3.97   0.000    -1.722167   -.5842197
-----------------+----------------------------------------------------------------
           alpha |   .3156273   .0916259                      .1786786    .5575408
----------------------------------------------------------------------------------

. 
. 
. 
. 
. 
. **Apendix** 
. 
. **Mil Models** 
. 
. **Model 16**
. 
. zinb cat1_sea_mil log_ngo_counts pts_mil gender_mil log_force_size log_pop_density log_gni_per_cap, inflate (
> log_ngo_counts pts_mil gender_mil log_force_size log_pop_density log_gni_per_cap) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -329.96013  
Iteration 1:  Log pseudolikelihood = -296.04561  
Iteration 2:  Log pseudolikelihood =  -278.7573  
Iteration 3:  Log pseudolikelihood = -275.59315  
Iteration 4:  Log pseudolikelihood = -274.51224  
Iteration 5:  Log pseudolikelihood = -274.04965  
Iteration 6:  Log pseudolikelihood = -273.57729  
Iteration 7:  Log pseudolikelihood = -273.56676  
Iteration 8:  Log pseudolikelihood = -273.56674  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -273.56674  (not concave)
Iteration 1:  Log pseudolikelihood = -260.18227  
Iteration 2:  Log pseudolikelihood = -235.72756  (not concave)
Iteration 3:  Log pseudolikelihood =  -233.2212  (not concave)
Iteration 4:  Log pseudolikelihood = -232.56137  (not concave)
Iteration 5:  Log pseudolikelihood =  -232.1999  (not concave)
Iteration 6:  Log pseudolikelihood = -232.10074  (not concave)
Iteration 7:  Log pseudolikelihood = -231.98414  
Iteration 8:  Log pseudolikelihood = -229.56642  (not concave)
Iteration 9:  Log pseudolikelihood = -229.24287  (not concave)
Iteration 10: Log pseudolikelihood = -229.07698  (not concave)
Iteration 11: Log pseudolikelihood = -228.96061  
Iteration 12: Log pseudolikelihood = -227.98163  
Iteration 13: Log pseudolikelihood = -227.14043  
Iteration 14: Log pseudolikelihood = -227.08826  
Iteration 15: Log pseudolikelihood = -227.08715  
Iteration 16: Log pseudolikelihood = -227.08715  

Zero-inflated negative binomial regression              Number of obs =    126
Inflation model: logit                                  Nonzero obs   =     71
                                                        Zero obs      =     55
                                                        Wald chi2(6)  = 208.17
Log pseudolikelihood = -227.0871                        Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 23 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
   cat1_sea_mil | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
cat1_sea_mil    |
 log_ngo_counts |  -.1913024   .4068986    -0.47   0.638    -.9888091    .6062042
    pts_mil_fil |   -.643367   .2879503    -2.23   0.025    -1.207739   -.0789949
     gender_mil |  -4.123468   4.136247    -1.00   0.319    -12.23036    3.983426
 log_force_size |   .5670873    .127052     4.46   0.000       .31807    .8161047
log_pop_density |  -.0456451   .0895715    -0.51   0.610     -.221202    .1299119
log_gni_per_cap |  -.7596067   .0951911    -7.98   0.000    -.9461778   -.5730356
          _cons |   5.543267   2.258798     2.45   0.014     1.116104     9.97043
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |  -1.513718    .721094    -2.10   0.036    -2.927037   -.1004001
    pts_mil_fil |    1.71382   1.372349     1.25   0.212    -.9759349    4.403574
     gender_mil |   30.63117   22.60458     1.36   0.175      -13.673    74.93533
 log_force_size |   -2.08545   .7976254    -2.61   0.009    -3.648768   -.5221333
log_pop_density |  -2.759119   1.204207    -2.29   0.022    -5.119321   -.3989175
log_gni_per_cap |   .7161933     .90642     0.79   0.429    -1.060357    2.492744
          _cons |   20.09769   6.779973     2.96   0.003     6.809188    33.38619
----------------+----------------------------------------------------------------
       /lnalpha |  -1.109443   .2435602    -4.56   0.000    -1.586812   -.6320735
----------------+----------------------------------------------------------------
          alpha |   .3297427   .0803122                      .2045768    .5314886
---------------------------------------------------------------------------------

. 
. 
. **Model 17**
. 
. zinb cat1_sea_mil log_ngo_counts_hr pts_mil gender_mil log_force_size log_gni_per_cap, inflate (log_ngo_count
> s_hr pts_mil gender_mil log_force_size log_gni_per_cap) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -329.96013  
Iteration 1:  Log pseudolikelihood = -295.94191  
Iteration 2:  Log pseudolikelihood = -280.11428  
Iteration 3:  Log pseudolikelihood = -278.02237  
Iteration 4:  Log pseudolikelihood = -277.25923  
Iteration 5:  Log pseudolikelihood = -277.13354  
Iteration 6:  Log pseudolikelihood = -277.13014  
Iteration 7:  Log pseudolikelihood = -277.13014  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -277.13014  (not concave)
Iteration 1:  Log pseudolikelihood = -258.35474  
Iteration 2:  Log pseudolikelihood = -240.81629  
Iteration 3:  Log pseudolikelihood = -238.75035  (not concave)
Iteration 4:  Log pseudolikelihood = -234.88392  
Iteration 5:  Log pseudolikelihood = -234.14484  
Iteration 6:  Log pseudolikelihood = -233.26044  
Iteration 7:  Log pseudolikelihood = -233.24724  
Iteration 8:  Log pseudolikelihood = -233.24721  
Iteration 9:  Log pseudolikelihood = -233.24721  

Zero-inflated negative binomial regression              Number of obs =    126
Inflation model: logit                                  Nonzero obs   =     71
                                                        Zero obs      =     55
                                                        Wald chi2(5)  = 135.84
Log pseudolikelihood = -233.2472                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 23 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
     cat1_sea_mil | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
cat1_sea_mil      |
log_ngo_counts_hr |  -.1346432   .4225682    -0.32   0.750    -.9628617    .6935752
      pts_mil_fil |  -.6442796   .2312747    -2.79   0.005     -1.09757   -.1909896
       gender_mil |   -4.82277   4.675871    -1.03   0.302    -13.98731    4.341769
   log_force_size |   .5931851   .1175432     5.05   0.000     .3628046    .8235655
  log_gni_per_cap |  -.7844279   .1003003    -7.82   0.000    -.9810128    -.587843
            _cons |   4.882423   1.799657     2.71   0.007      1.35516    8.409687
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |  -2.155878    .886038    -2.43   0.015     -3.89248   -.4192753
      pts_mil_fil |   .1741856   2.078937     0.08   0.933    -3.900455    4.248827
       gender_mil |   22.62823   14.73047     1.54   0.125    -6.242962    51.49942
   log_force_size |  -.6635662   .5925894    -1.12   0.263     -1.82502    .4978877
  log_gni_per_cap |  -.2656302   .7583877    -0.35   0.726    -1.752043    1.220782
            _cons |   12.04493   9.082728     1.33   0.185    -5.756886    29.84675
------------------+----------------------------------------------------------------
         /lnalpha |  -1.061013   .2607693    -4.07   0.000    -1.572112   -.5499146
------------------+----------------------------------------------------------------
            alpha |    .346105   .0902536                      .2076063    .5769991
-----------------------------------------------------------------------------------

. 
. **Model 18**
. 
. zinb sea_mil log_ngo_counts_w pts_mil  gender_mil log_force_size, inflate (log_ngo_counts_w pts_mil gender_mi
> l log_force_size ) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -249.72027  (not concave)
Iteration 1:  Log pseudolikelihood = -203.34526  
Iteration 2:  Log pseudolikelihood = -195.50171  
Iteration 3:  Log pseudolikelihood = -191.84152  
Iteration 4:  Log pseudolikelihood = -191.51829  
Iteration 5:  Log pseudolikelihood = -191.51048  
Iteration 6:  Log pseudolikelihood = -191.51047  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -191.51047  (not concave)
Iteration 1:  Log pseudolikelihood = -182.18548  (not concave)
Iteration 2:  Log pseudolikelihood = -176.55599  
Iteration 3:  Log pseudolikelihood = -172.46128  
Iteration 4:  Log pseudolikelihood = -170.50811  
Iteration 5:  Log pseudolikelihood = -169.52067  
Iteration 6:  Log pseudolikelihood = -169.29242  
Iteration 7:  Log pseudolikelihood = -169.27237  
Iteration 8:  Log pseudolikelihood =  -169.2716  
Iteration 9:  Log pseudolikelihood =  -169.2716  

Zero-inflated negative binomial regression              Number of obs =    144
Inflation model: logit                                  Nonzero obs   =     49
                                                        Zero obs      =     95
                                                        Wald chi2(4)  =  45.46
Log pseudolikelihood = -169.2716                        Prob > chi2   = 0.0000

                                   (Std. err. adjusted for 27 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
         sea_mil | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_mil          |
log_ngo_counts_w |  -1.421915   1.230313    -1.16   0.248    -3.833283    .9894539
     pts_mil_fil |  -.4969558   .3954178    -1.26   0.209     -1.27196    .2780489
      gender_mil |    .294227   12.39341     0.02   0.981     -23.9964    24.58486
  log_force_size |   1.804469   .7753104     2.33   0.020     .2848881    3.324049
           _cons |  -9.255263   3.061103    -3.02   0.002    -15.25491   -3.255612
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |  -7.605201   12.72038    -0.60   0.550    -32.53669    17.32629
     pts_mil_fil |  -5.935357   12.94774    -0.46   0.647    -31.31245    19.44174
      gender_mil |    53.5368   62.86346     0.85   0.394    -69.67331    176.7469
  log_force_size |   6.106107   11.51872     0.53   0.596    -16.47017    28.68238
           _cons |  -18.74348   35.79204    -0.52   0.601    -88.89459    51.40763
-----------------+----------------------------------------------------------------
        /lnalpha |  -.4513276   1.285581    -0.35   0.726    -2.971021    2.068366
-----------------+----------------------------------------------------------------
           alpha |   .6367822   .8186353                       .051251    7.911881
----------------------------------------------------------------------------------

. 
. **pol Models**
. 
. **Model 19**
. 
. zinb cat1_sea_pol log_ngo_counts pts_mission_pol gender_pol log_force_size log_pop_density , inflate (log_ngo
> _counts pts_mission_pol gender_pol log_force_size log_pop_density) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -196.46058  (not concave)
Iteration 1:  Log pseudolikelihood = -184.60661  
Iteration 2:  Log pseudolikelihood = -165.00968  
Iteration 3:  Log pseudolikelihood = -164.15343  
Iteration 4:  Log pseudolikelihood = -164.11398  
Iteration 5:  Log pseudolikelihood = -164.11384  
Iteration 6:  Log pseudolikelihood = -164.11384  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -164.11384  
Iteration 1:  Log pseudolikelihood = -154.33808  
Iteration 2:  Log pseudolikelihood = -150.46187  
Iteration 3:  Log pseudolikelihood = -150.20143  
Iteration 4:  Log pseudolikelihood = -150.19204  
Iteration 5:  Log pseudolikelihood = -150.19199  
Iteration 6:  Log pseudolikelihood = -150.19199  

Zero-inflated negative binomial regression              Number of obs =    101
Inflation model: logit                                  Nonzero obs   =     46
                                                        Zero obs      =     55
                                                        Wald chi2(5)  =  71.20
Log pseudolikelihood = -150.192                         Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 21 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
   cat1_sea_pol | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
cat1_sea_pol    |
 log_ngo_counts |  -.8494347   .3311071    -2.57   0.010    -1.498393   -.2004766
pts_mission_pol |  -.4746725   .4158965    -1.14   0.254    -1.289815    .3404697
     gender_pol |   5.782578   3.382912     1.71   0.087    -.8478083    12.41296
 log_force_size |     .77638   .1812685     4.28   0.000     .4211004     1.13166
log_pop_density |   .4764049   .1248493     3.82   0.000     .2317049     .721105
          _cons |  -2.855406   1.630806    -1.75   0.080    -6.051726    .3409151
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |   .5277714   2.151829     0.25   0.806    -3.689736    4.745279
pts_mission_pol |  -1.536167   2.113224    -0.73   0.467    -5.678009    2.605676
     gender_pol |   7.796272   8.732896     0.89   0.372    -9.319889    24.91243
 log_force_size |  -.2699438   1.453217    -0.19   0.853    -3.118197     2.57831
log_pop_density |  -.9714388   .4965513    -1.96   0.050    -1.944661    .0017838
          _cons |   4.600676   6.661915     0.69   0.490    -8.456438    17.65779
----------------+----------------------------------------------------------------
       /lnalpha |  -.6780924   .7180807    -0.94   0.345    -2.085505    .7293198
----------------+----------------------------------------------------------------
          alpha |   .5075843   .3644865                      .1242444     2.07367
---------------------------------------------------------------------------------

. 
. **Model 20**
. 
. zinb cat1_sea_pol log_ngo_counts_hr pts_mission_pol  log_force_size, inflate (log_ngo_counts_hr pts_mission_p
> ol  log_force_size) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -262.14844  (not concave)
Iteration 1:  Log pseudolikelihood = -245.08679  (not concave)
Iteration 2:  Log pseudolikelihood = -225.06903  
Iteration 3:  Log pseudolikelihood =  -221.9648  
Iteration 4:  Log pseudolikelihood = -221.71059  
Iteration 5:  Log pseudolikelihood = -221.70257  
Iteration 6:  Log pseudolikelihood = -221.70257  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -221.70257  
Iteration 1:  Log pseudolikelihood = -215.84843  
Iteration 2:  Log pseudolikelihood = -212.84522  
Iteration 3:  Log pseudolikelihood = -209.29214  
Iteration 4:  Log pseudolikelihood = -207.61321  
Iteration 5:  Log pseudolikelihood = -207.07163  
Iteration 6:  Log pseudolikelihood = -206.85257  
Iteration 7:  Log pseudolikelihood = -206.78009  
Iteration 8:  Log pseudolikelihood = -206.75246  
Iteration 9:  Log pseudolikelihood = -206.63859  
Iteration 10: Log pseudolikelihood = -206.62386  
Iteration 11: Log pseudolikelihood = -206.62036  
Iteration 12: Log pseudolikelihood = -206.61986  
Iteration 13: Log pseudolikelihood = -206.61979  
Iteration 14: Log pseudolikelihood = -206.61978  

Zero-inflated negative binomial regression                 Number of obs = 134
Inflation model: logit                                     Nonzero obs   =  61
                                                           Zero obs      =  73
                                                           Wald chi2(3)  =   .
Log pseudolikelihood = -206.6198                           Prob > chi2   =   .

                                    (Std. err. adjusted for 24 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
     cat1_sea_pol | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
cat1_sea_pol      |
log_ngo_counts_hr |  -1.665965   .3347849    -4.98   0.000    -2.322132   -1.009799
  pts_mission_pol |   .1454621   .2070508     0.70   0.482      -.26035    .5512742
   log_force_size |   .7475184   .2236539     3.34   0.001     .3091648    1.185872
            _cons |   .4273347   1.312772     0.33   0.745    -2.145651     3.00032
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |  -145.3854   18.64964    -7.80   0.000     -181.938   -108.8328
  pts_mission_pol |   89.35837   12.68414     7.04   0.000     64.49791    114.2188
   log_force_size |  -67.99361    8.48542    -8.01   0.000    -84.62473   -51.36249
            _cons |   606.4988   75.44013     8.04   0.000     458.6389    754.3588
------------------+----------------------------------------------------------------
         /lnalpha |   .1710847   .2867975     0.60   0.551     -.391028    .7331974
------------------+----------------------------------------------------------------
            alpha |   1.186591   .3403113                      .6763612    2.081726
-----------------------------------------------------------------------------------

. 
. **Model 21**
. 
. zinb sea_pol log_ngo_counts_w pts_mission_pol   log_force_size log_pop_density, inflate (log_ngo_counts_w pts
> _mission_pol  log_force_size log_pop_density) difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -174.47156  (not concave)
Iteration 1:  Log pseudolikelihood = -159.98861  (not concave)
Iteration 2:  Log pseudolikelihood = -147.05909  (not concave)
Iteration 3:  Log pseudolikelihood =  -140.8735  
Iteration 4:  Log pseudolikelihood = -134.53525  
Iteration 5:  Log pseudolikelihood = -133.12368  
Iteration 6:  Log pseudolikelihood = -132.91802  
Iteration 7:  Log pseudolikelihood = -132.91686  
Iteration 8:  Log pseudolikelihood = -132.91686  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -132.91686  
Iteration 1:  Log pseudolikelihood = -124.95169  
Iteration 2:  Log pseudolikelihood = -122.72605  
Iteration 3:  Log pseudolikelihood = -121.72852  
Iteration 4:  Log pseudolikelihood = -120.19715  
Iteration 5:  Log pseudolikelihood = -119.55245  
Iteration 6:  Log pseudolikelihood = -119.29328  
Iteration 7:  Log pseudolikelihood = -119.19723  (not concave)
Iteration 8:  Log pseudolikelihood = -118.31917  
Iteration 9:  Log pseudolikelihood = -118.06634  (backed up)
Iteration 10: Log pseudolikelihood = -116.34356  
Iteration 11: Log pseudolikelihood = -115.41872  
Iteration 12: Log pseudolikelihood = -114.48729  
Iteration 13: Log pseudolikelihood = -114.27544  
Iteration 14: Log pseudolikelihood =  -114.2247  
Iteration 15: Log pseudolikelihood = -114.21561  
Iteration 16: Log pseudolikelihood =   -114.215  
Iteration 17: Log pseudolikelihood = -114.21499  

Zero-inflated negative binomial regression              Number of obs =    134
Inflation model: logit                                  Nonzero obs   =     46
                                                        Zero obs      =     88
                                                        Wald chi2(4)  = 131.47
Log pseudolikelihood = -114.215                         Prob > chi2   = 0.0000

                                   (Std. err. adjusted for 24 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
         sea_pol | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_pol          |
log_ngo_counts_w |  -.7233955    .179613    -4.03   0.000    -1.075431   -.3713605
 pts_mission_pol |   .0899874   .4493567     0.20   0.841    -.7907355    .9707103
  log_force_size |   1.102352   .2228119     4.95   0.000     .6656487    1.539055
 log_pop_density |   .5131954   .1304572     3.93   0.000     .2575041    .7688867
           _cons |  -9.716191   2.047622    -4.75   0.000    -13.72946   -5.702926
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |   196.6597   132.0915     1.49   0.137    -62.23488    455.5543
 pts_mission_pol |  -38.53942   22.50469    -1.71   0.087     -82.6478     5.56896
  log_force_size |   28.09649   16.21757     1.73   0.083    -3.689363    59.88234
 log_pop_density |   9.633926   6.956215     1.38   0.166    -4.000006    23.26786
           _cons |  -850.3743   549.6653    -1.55   0.122    -1927.699    226.9498
-----------------+----------------------------------------------------------------
        /lnalpha |  -1.507677    1.34969    -1.12   0.264    -4.153021    1.137668
-----------------+----------------------------------------------------------------
           alpha |   .2214238   .2988536                      .0157169    3.119485
----------------------------------------------------------------------------------

. 
. ********************
. *** Figures ***
. ********************
. 
. ********************
. *** Figure 3 ***
. ********************
. 
. *** Run model and save estimates ***
. eststo m1: zinb sea_cat1_total log_ngo_counts pts_mission gender_ratio log_force_size log_pop_density log_gni
> _per_cap, inflate (log_ngo_counts pts_mission gender_ratio log_force_size log_pop_density log_gni_per_cap) cl
> uster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -419.27563  
Iteration 2:  Log pseudolikelihood = -409.54704  
Iteration 3:  Log pseudolikelihood = -408.73965  
Iteration 4:  Log pseudolikelihood = -408.71301  
Iteration 5:  Log pseudolikelihood = -408.71293  
Iteration 6:  Log pseudolikelihood = -408.71293  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -408.71293  (not concave)
Iteration 1:  Log pseudolikelihood = -390.95997  
Iteration 2:  Log pseudolikelihood = -361.74367  
Iteration 3:  Log pseudolikelihood = -347.12438  
Iteration 4:  Log pseudolikelihood = -345.00778  
Iteration 5:  Log pseudolikelihood = -344.94993  
Iteration 6:  Log pseudolikelihood = -344.94956  
Iteration 7:  Log pseudolikelihood = -344.94955  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(6)  = 118.78
Log pseudolikelihood = -344.9496                        Prob > chi2   = 0.0000

                                  (Std. err. adjusted for 27 clusters in mission)
---------------------------------------------------------------------------------
                |               Robust
 sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
----------------+----------------------------------------------------------------
sea_cat1_total  |
 log_ngo_counts |    -.36833   .1474205    -2.50   0.012     -.657269   -.0793911
    pts_mission |   .0574619   .1876215     0.31   0.759    -.3102696    .4251933
   gender_ratio |  -5.078576    4.50021    -1.13   0.259    -13.89883    3.741675
 log_force_size |   .4288427   .0835315     5.13   0.000     .2651239    .5925614
log_pop_density |   .1363278   .0765101     1.78   0.075    -.0136291    .2862848
log_gni_per_cap |  -.6591225   .1302719    -5.06   0.000    -.9144507   -.4037943
          _cons |   4.385876   1.692811     2.59   0.010     1.068028    7.703723
----------------+----------------------------------------------------------------
inflate         |
 log_ngo_counts |   .4239687   .6751336     0.63   0.530    -.8992688    1.747206
    pts_mission |  -.5809327   .8246916    -0.70   0.481    -2.197299    1.035433
   gender_ratio |  -3.197862   19.99518    -0.16   0.873    -42.38769    35.99196
 log_force_size |  -1.027214   .2963761    -3.47   0.001    -1.608101   -.4463279
log_pop_density |   -1.15948   .5723608    -2.03   0.043    -2.281287   -.0376736
log_gni_per_cap |  -.4634157   .4543376    -1.02   0.308    -1.353901    .4270697
          _cons |   12.88599   4.944083     2.61   0.009     3.195768    22.57622
----------------+----------------------------------------------------------------
       /lnalpha |  -1.020606   .2249257    -4.54   0.000    -1.461453   -.5797603
----------------+----------------------------------------------------------------
          alpha |   .3603763   .0810579                      .2318992    .5600326
---------------------------------------------------------------------------------

. 
. *** Loop to set global macros that hold natural logged counts of NGOs ***
. * Since range is 7 - 330, I will do 10 - 330
. macro drop _all

. forvalues i = 1(1)33 {
  2.         global set_`i' = ln(`i'0)
  3. }

. 
. 
. *** Estimate predicted counts based on set values above ***
. estimates restore m1
(results m1 are active now)

. margins, at(log_ngo_counts = ($set_1 $set_2 $set_3 $set_4 $set_5 $set_6 $set_7 $set_8 $set_9 $set_10 $set_11 
> $set_12 $set_13 $set_14 $set_15 $set_16 $set_17 $set_18 $set_19 $set_20 $set_21 $set_22 $set_23 $set_24 $set_
> 25 $set_26 $set_27 $set_28 $set_29 $set_30 $set_31 $set_32 $set_33 )) atmeans post saving(mars, replace)

Adjusted predictions                                       Number of obs = 165
Model VCE: Robust

Expression: Predicted number of events, predict()
1._at:  log_ngo_counts  = 2.302585
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
2._at:  log_ngo_counts  = 2.995732
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
3._at:  log_ngo_counts  = 3.401197
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
4._at:  log_ngo_counts  = 3.688879
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
5._at:  log_ngo_counts  = 3.912023
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
6._at:  log_ngo_counts  = 4.094345
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
7._at:  log_ngo_counts  = 4.248495
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
8._at:  log_ngo_counts  = 4.382027
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
9._at:  log_ngo_counts  =  4.49981
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
10._at: log_ngo_counts  =  4.60517
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
11._at: log_ngo_counts  =  4.70048
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
12._at: log_ngo_counts  = 4.787492
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
13._at: log_ngo_counts  = 4.867534
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
14._at: log_ngo_counts  = 4.941642
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
15._at: log_ngo_counts  = 5.010635
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
16._at: log_ngo_counts  = 5.075174
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
17._at: log_ngo_counts  = 5.135798
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
18._at: log_ngo_counts  = 5.192957
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
19._at: log_ngo_counts  = 5.247024
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
20._at: log_ngo_counts  = 5.298317
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
21._at: log_ngo_counts  = 5.347108
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
22._at: log_ngo_counts  = 5.393628
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
23._at: log_ngo_counts  = 5.438079
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
24._at: log_ngo_counts  = 5.480639
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
25._at: log_ngo_counts  = 5.521461
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
26._at: log_ngo_counts  = 5.560682
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
27._at: log_ngo_counts  = 5.598422
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
28._at: log_ngo_counts  =  5.63479
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
29._at: log_ngo_counts  = 5.669881
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
30._at: log_ngo_counts  = 5.703782
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
31._at: log_ngo_counts  = 5.736572
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
32._at: log_ngo_counts  = 5.768321
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)
33._at: log_ngo_counts  = 5.799093
        pts_mission     = 2.012993 (mean)
        gender_ratio    = .0463783 (mean)
        log_force_size  = 7.011212 (mean)
        log_pop_density = 4.317207 (mean)
        log_gni_per_cap = 7.485803 (mean)

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
         _at |
          1  |   7.663022   2.729158     2.81   0.005     2.313971    13.01207
          2  |   5.852815   1.625999     3.60   0.000     2.665915    9.039714
          3  |   4.989131   1.190334     4.19   0.000     2.656119    7.322144
          4  |    4.45001   .9576307     4.65   0.000     2.573088    6.326931
          5  |   4.069494   .8156224     4.99   0.000     2.470903    5.668084
          6  |   3.781043   .7222707     5.23   0.000     2.365419    5.196668
          7  |   3.551885   .6579552     5.40   0.000     2.262316    4.841453
          8  |   3.363681    .612197     5.49   0.000     2.163797    4.563565
          9  |   3.205235   .5788724     5.54   0.000     2.070666    4.339804
         10  |   3.069253   .5541696     5.54   0.000     1.983101    4.155406
         11  |   2.950749   .5356046     5.51   0.000     1.900983    4.000514
         12  |   2.846173   .5215016     5.46   0.000     1.824049    3.868297
         13  |   2.752922   .5106991     5.39   0.000      1.75197    3.753873
         14  |   2.669033    .502374     5.31   0.000     1.684398    3.653668
         15  |   2.592995   .4959324     5.23   0.000     1.620985    3.565004
         16  |    2.52362   .4909389     5.14   0.000     1.561398    3.485843
         17  |   2.459962   .4870696     5.05   0.000     1.505324    3.414601
         18  |   2.401255    .484081     4.96   0.000     1.452474    3.350036
         19  |    2.34687   .4817876     4.87   0.000     1.402584    3.291156
         20  |   2.296286   .4800468     4.78   0.000     1.355412    3.237161
         21  |   2.249069    .478748     4.70   0.000      1.31074    3.187397
         22  |   2.204848   .4778044     4.61   0.000     1.268369    3.141328
         23  |   2.163313   .4771479     4.53   0.000      1.22812    3.098505
         24  |   2.124192   .4767242     4.46   0.000      1.18983    3.058555
         25  |   2.087256   .4764896     4.38   0.000     1.153353    3.021158
         26  |   2.052301   .4764095     4.31   0.000     1.118555    2.986046
         27  |   2.019152   .4764554     4.24   0.000     1.085316    2.952987
         28  |   1.987654   .4766043     4.17   0.000     1.053527    2.921781
         29  |    1.95767   .4768374     4.11   0.000     1.023086    2.892254
         30  |    1.92908   .4771393     4.04   0.000     .9939038    2.864256
         31  |   1.901775   .4774973     3.98   0.000     .9658979    2.837653
         32  |    1.87566   .4779007     3.92   0.000      .938992    2.812328
         33  |   1.850648   .4783408     3.87   0.000     .9131173    2.788179
------------------------------------------------------------------------------

. 
. 
. *** Import predicted values *** 
. use mars.dta, clear
(Created by command margins; also see char list)

. 
. 
. *** Keep point estimates, upper bound, and lower bound ***
. keep _margin _ci_lb _ci_ub

. 
. 
. *** Generate counts of NGOs for the graph ***
. gen ngo_counts = [_n] * 10

. 
. 
. *** Make a graph ***
. graph twoway line _margin ngo_count, clwidth(medium) clcolor(black) clpattern(solid) sort || line _ci_lb ngo_
> count, clpattern(dash) clwidth(thin) clcolor(black) sort || line _ci_ub ngo_count, clpattern(dash) clwidth(th
> in) clcolor(black) sort legend(off) xlabel(10(40)330) xtitle("Number of NGOs", size(medsmall)) ytitle("Predic
> ted Count of Category 1 and SEA Misconduct", size(medsmall)) scheme(sj) title("Effect of NGO Counts on Catego
> ry 1 and SEA Misconduct", size(medlarge))

. graph export "Model_1.jpg", as(jpg) name("Graph") quality(90) replace
file /Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Model_1.jpg saved as JPG format

. 
. 
. ********************
. *** Figure 4 ***
. ********************
. 
. *** Import pkat .csv file ***
. import delimited "Final Data.csv", clear  
(encoding automatically selected: ISO-8859-1)
(121 vars, 208 obs)

. 
. 
. *** Generate logged variables ***
. gen log_force_size = ln(force_size)
(19 missing values generated)

. gen log_pop_density = ln(pop_density)
(18 missing values generated)

. gen log_gni_per_cap = ln(gni_per_cap)
(26 missing values generated)

. gen log_gdp_mission = ln(gdp_mission)
(19 missing values generated)

. 
. 
. *** Run model and save estimates ***
. eststo m1: zinb sea_cat1_total log_ngo_counts_hr pts_mission gender_ratio log_force_size log_gni_per_cap, inf
> late (log_ngo_counts_hr pts_mission gender_ratio log_force_size log_gni_per_cap) cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -475.71759  
Iteration 1:  Log pseudolikelihood = -419.65931  
Iteration 2:  Log pseudolikelihood = -411.02426  
Iteration 3:  Log pseudolikelihood = -410.64733  
Iteration 4:  Log pseudolikelihood = -410.64209  
Iteration 5:  Log pseudolikelihood = -410.64208  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -410.64208  (not concave)
Iteration 1:  Log pseudolikelihood = -392.81625  
Iteration 2:  Log pseudolikelihood = -357.44927  (backed up)
Iteration 3:  Log pseudolikelihood = -350.37621  
Iteration 4:  Log pseudolikelihood = -349.95929  
Iteration 5:  Log pseudolikelihood = -349.95573  
Iteration 6:  Log pseudolikelihood = -349.95573  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     99
                                                        Zero obs      =     66
                                                        Wald chi2(5)  = 153.06
Log pseudolikelihood = -349.9557                        Prob > chi2   = 0.0000

                                    (Std. err. adjusted for 27 clusters in mission)
-----------------------------------------------------------------------------------
                  |               Robust
   sea_cat1_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
------------------+----------------------------------------------------------------
sea_cat1_total    |
log_ngo_counts_hr |  -.4241099   .2004736    -2.12   0.034    -.8170308   -.0311889
      pts_mission |  -.0261804   .2010457    -0.13   0.896    -.4202228    .3678619
     gender_ratio |   -4.18209   4.209002    -0.99   0.320    -12.43158    4.067403
   log_force_size |   .4345677   .0828233     5.25   0.000      .272237    .5968983
  log_gni_per_cap |  -.5981868   .1166087    -5.13   0.000    -.8267356    -.369638
            _cons |   4.453133   1.555192     2.86   0.004     1.405013    7.501253
------------------+----------------------------------------------------------------
inflate           |
log_ngo_counts_hr |    .165482   .5491179     0.30   0.763    -.9107692    1.241733
      pts_mission |  -1.100595   .8386933    -1.31   0.189    -2.744404    .5432134
     gender_ratio |  -3.677691   13.14812    -0.28   0.780    -29.44753    22.09215
   log_force_size |   -.635038   .2166404    -2.93   0.003    -1.059645   -.2104305
  log_gni_per_cap |  -.6728958   .3678973    -1.83   0.067    -1.393961    .0481697
            _cons |   9.588462   5.184891     1.85   0.064    -.5737382    19.75066
------------------+----------------------------------------------------------------
         /lnalpha |  -.9153787   .2015872    -4.54   0.000    -1.310482   -.5202751
------------------+----------------------------------------------------------------
            alpha |    .400365   .0807084                      .2696899     .594357
-----------------------------------------------------------------------------------

. 
. *** Loop to set global macros that hold natural logged counts of NGOs ***
. * Since range is 9 - 97, I'll do 10 - 100 
. macro drop _all

. forvalues i = 10(5)100 {
  2.         global set_`i' = ln(`i')
  3. }

. 
. *** Estimate predicted counts based on set values above ***
. estimates restore m1
(results m1 are active now)

. margins, at(log_ngo_counts = ($set_10 $set_15 $set_20 $set_25 $set_30 $set_35 $set_40 $set_45 $set_50 $set_55
>  $set_60 $set_65 $set_70 $set_75 $set_80 $set_85 $set_90 $set_95 $set_100 )) atmeans post saving(mars2, repla
> ce)

Adjusted predictions                                       Number of obs = 165
Model VCE: Robust

Expression: Predicted number of events, predict()
1._at:  log_ngo_counts~r = 2.302585
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
2._at:  log_ngo_counts~r =  2.70805
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
3._at:  log_ngo_counts~r = 2.995732
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
4._at:  log_ngo_counts~r = 3.218876
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
5._at:  log_ngo_counts~r = 3.401197
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
6._at:  log_ngo_counts~r = 3.555348
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
7._at:  log_ngo_counts~r = 3.688879
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
8._at:  log_ngo_counts~r = 3.806662
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
9._at:  log_ngo_counts~r = 3.912023
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
10._at: log_ngo_counts~r = 4.007333
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
11._at: log_ngo_counts~r = 4.094345
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
12._at: log_ngo_counts~r = 4.174387
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
13._at: log_ngo_counts~r = 4.248495
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
14._at: log_ngo_counts~r = 4.317488
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
15._at: log_ngo_counts~r = 4.382027
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
16._at: log_ngo_counts~r = 4.442651
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
17._at: log_ngo_counts~r =  4.49981
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
18._at: log_ngo_counts~r = 4.553877
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)
19._at: log_ngo_counts~r =  4.60517
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_gni_per_cap  = 7.485803 (mean)

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
         _at |
          1  |   5.260723   1.853807     2.84   0.005     1.627328    8.894117
          2  |   4.390158     1.2503     3.51   0.000     1.939614    6.840701
          3  |    3.86009   .9345612     4.13   0.000     2.028384    5.691797
          4  |    3.49276   .7481418     4.67   0.000     2.026429    4.959091
          5  |   3.218323   .6322117     5.09   0.000     1.979211    4.457435
          6  |   3.002909   .5589607     5.37   0.000     1.907367    4.098452
          7  |   2.827805   .5129767     5.51   0.000      1.82239    3.833221
          8  |   2.681703   .4847986     5.53   0.000     1.731515    3.631891
          9  |   2.557308   .4682974     5.46   0.000     1.639462    3.475154
         10  |    2.44967   .4594222     5.33   0.000     1.549219    3.350121
         11  |   2.355295   .4554886     5.17   0.000     1.462553    3.248036
         12  |   2.271634   .4547227     5.00   0.000     1.380394    3.162874
         13  |    2.19678   .4559554     4.82   0.000     1.303124    3.090436
         14  |   2.129271   .4584176     4.64   0.000     1.230789    3.027752
         15  |   2.067964   .4616033     4.48   0.000     1.163238     2.97269
         16  |   2.011955   .4651801     4.33   0.000     1.100218    2.923691
         17  |   1.960512   .4689303     4.18   0.000     1.041425    2.879598
         18  |   1.913039   .4727119     4.05   0.000     .9865406    2.839537
         19  |   1.869044   .4764337     3.92   0.000     .9352515    2.802837
------------------------------------------------------------------------------

. 
. 
. *** Import predicted values *** 
. use mars2.dta, clear
(Created by command margins; also see char list)

. 
. 
. *** Keep point estimates, upper bound, and lower bound ***
. keep _margin _ci_lb _ci_ub

. 
. 
. *** Generate counts of NGOs for the graph ***
. gen ngo_counts = 10

. replace ngo_counts = 15 in 2
(1 real change made)

. replace ngo_counts = 20 in 3
(1 real change made)

. replace ngo_counts = 25 in 4
(1 real change made)

. replace ngo_counts = 30 in 5
(1 real change made)

. replace ngo_counts = 35 in 6
(1 real change made)

. replace ngo_counts = 40 in 7
(1 real change made)

. replace ngo_counts = 45 in 8
(1 real change made)

. replace ngo_counts = 50 in 9
(1 real change made)

. replace ngo_counts = 55 in 10
(1 real change made)

. replace ngo_counts = 60 in 11
(1 real change made)

. replace ngo_counts = 65 in 12
(1 real change made)

. replace ngo_counts = 70 in 13
(1 real change made)

. replace ngo_counts = 75 in 14
(1 real change made)

. replace ngo_counts = 80 in 15
(1 real change made)

. replace ngo_counts = 85 in 16
(1 real change made)

. replace ngo_counts = 90 in 17
(1 real change made)

. replace ngo_counts = 95 in 18
(1 real change made)

. replace ngo_counts = 100 in 19
(1 real change made)

. 
. 
. *** Make a graph ***
. graph twoway line _margin ngo_counts, clwidth(medium) clcolor(black) clpattern(solid) sort || line _ci_lb ngo
> _counts, clpattern(dash) clwidth(thin) clcolor(black) sort || line _ci_ub ngo_counts, clpattern(dash) clwidth
> (thin) clcolor(black) sort legend(off) xlabel(10(10)100) xtitle("Number of Human Rights NGOs", size(medsmall)
> ) ytitle("Predicted Count of Category 1 and SEA Misconduct", size(medsmall)) scheme(sj) title("Effect of Huma
> n Rights NGO Counts on Category 1 and SEA Misconduct", size(medium))

. graph export "Model_7.jpg", as(jpg) name("Graph") quality(90) replace
file /Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Model_7.jpg saved as JPG format

. 
. 
. *********************
. *** Figure 5 ***
. *********************
. 
. *** Import pkat .csv file ***
. import delimited "Final Data.csv", clear  
(encoding automatically selected: ISO-8859-1)
(121 vars, 208 obs)

. 
. 
. *** Generate logged variables ***
. gen log_force_size = ln(force_size)
(19 missing values generated)

. gen log_pop_density = ln(pop_density)
(18 missing values generated)

. gen log_gni_per_cap = ln(gni_per_cap)
(26 missing values generated)

. gen log_gdp_mission = ln(gdp_mission)
(19 missing values generated)

. 
. 
. *** Run model and save estimates ***
. *
. * 
. *
. eststo m1: zinb sea_total log_ngo_counts_w pts_mission  gender_ratio log_force_size log_pop_density log_gni_p
> er_cap , inflate (log_ngo_counts_w pts_mission gender_ratio log_force_size log_pop_density log_gni_per_cap  )
>  difficult cluster(mission)

Fitting constant-only model:

Iteration 0:  Log pseudolikelihood = -338.20938  (not concave)
Iteration 1:  Log pseudolikelihood = -271.33666  
Iteration 2:  Log pseudolikelihood = -269.45494  
Iteration 3:  Log pseudolikelihood =  -269.3951  
Iteration 4:  Log pseudolikelihood = -269.39493  
Iteration 5:  Log pseudolikelihood = -269.39493  

Fitting full model:

Iteration 0:  Log pseudolikelihood = -269.39493  (not concave)
Iteration 1:  Log pseudolikelihood = -246.52504  (not concave)
Iteration 2:  Log pseudolikelihood = -230.95256  (not concave)
Iteration 3:  Log pseudolikelihood =   -224.612  
Iteration 4:  Log pseudolikelihood = -220.78869  
Iteration 5:  Log pseudolikelihood = -220.07508  
Iteration 6:  Log pseudolikelihood = -219.78068  
Iteration 7:  Log pseudolikelihood = -219.68811  
Iteration 8:  Log pseudolikelihood = -219.68705  
Iteration 9:  Log pseudolikelihood = -219.68704  

Zero-inflated negative binomial regression              Number of obs =    165
Inflation model: logit                                  Nonzero obs   =     69
                                                        Zero obs      =     96
                                                        Wald chi2(6)  =  77.98
Log pseudolikelihood = -219.687                         Prob > chi2   = 0.0000

                                   (Std. err. adjusted for 27 clusters in mission)
----------------------------------------------------------------------------------
                 |               Robust
       sea_total | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-----------------+----------------------------------------------------------------
sea_total        |
log_ngo_counts_w |  -.2746243   .2240927    -1.23   0.220     -.713838    .1645893
     pts_mission |    .291087   .2966418     0.98   0.326    -.2903203    .8724943
    gender_ratio |  -4.294218   10.52344    -0.41   0.683    -24.91978    16.33134
  log_force_size |   .3128914   .2282017     1.37   0.170    -.1343757    .7601586
 log_pop_density |   .1786824   .1599524     1.12   0.264    -.1348186    .4921834
 log_gni_per_cap |  -1.234051   .2633893    -4.69   0.000    -1.750284   -.7178172
           _cons |   6.578112   2.763101     2.38   0.017     1.162534    11.99369
-----------------+----------------------------------------------------------------
inflate          |
log_ngo_counts_w |   2.281688   4.967398     0.46   0.646    -7.454234    12.01761
     pts_mission |   1.574281   1.467459     1.07   0.283    -1.301886    4.450448
    gender_ratio |   42.78145   69.65718     0.61   0.539    -93.74411     179.307
  log_force_size |  -2.469222   1.043094    -2.37   0.018    -4.513649   -.4247953
 log_pop_density |  -3.418284    2.30704    -1.48   0.138        -7.94    1.103431
 log_gni_per_cap |  -1.637711   1.391216    -1.18   0.239    -4.364444    1.089022
           _cons |   28.34446   19.25637     1.47   0.141    -9.397339    66.08625
-----------------+----------------------------------------------------------------
        /lnalpha |  -1.141814   .3098731    -3.68   0.000    -1.749154   -.5344742
-----------------+----------------------------------------------------------------
           alpha |   .3192393   .0989237                       .173921    .5859773
----------------------------------------------------------------------------------

. 
. 
. *** Loop to set global macros that hold natural logged counts of NGOs ***
. * Since range is 1 - 49, I'll do 1 - 50
. macro drop _all

. forvalues i = 5(5)50 {
  2.         global set_`i' = ln(`i')
  3. }

. global set_1 = ln(1)

. 
. 
. *** Estimate predicted counts based on set values above ***
. estimates restore m1
(results m1 are active now)

. margins, at(log_ngo_counts = ($set_1 $set_5 $set_10 $set_15 $set_20 $set_25 $set_30 $set_35 $set_40 $set_45 $
> set_50 )) atmeans post saving(mars3, replace)

Adjusted predictions                                       Number of obs = 165
Model VCE: Robust

Expression: Predicted number of events, predict()
1._at:  log_ngo_counts_w =        0
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
2._at:  log_ngo_counts_w = 1.609438
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
3._at:  log_ngo_counts_w = 2.302585
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
4._at:  log_ngo_counts_w =  2.70805
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
5._at:  log_ngo_counts_w = 2.995732
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
6._at:  log_ngo_counts_w = 3.218876
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
7._at:  log_ngo_counts_w = 3.401197
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
8._at:  log_ngo_counts_w = 3.555348
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
9._at:  log_ngo_counts_w = 3.688879
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
10._at: log_ngo_counts_w = 3.806662
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)
11._at: log_ngo_counts_w = 3.912023
        pts_mission      = 2.012993 (mean)
        gender_ratio     = .0463783 (mean)
        log_force_size   = 7.011212 (mean)
        log_pop_density  = 4.317207 (mean)
        log_gni_per_cap  = 7.485803 (mean)

------------------------------------------------------------------------------
             |            Delta-method
             |     Margin   std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
         _at |
          1  |    1.99808   .7081089     2.82   0.005     .6102121    3.385948
          2  |   1.283305   .4885173     2.63   0.009     .3258284    2.240781
          3  |    1.05769     .51276     2.06   0.039     .0526988    2.062681
          4  |   .9408417   .5184005     1.81   0.070    -.0752047    1.956888
          5  |   .8618185   .5065893     1.70   0.089    -.1310782    1.854715
          6  |   .8009926   .4799914     1.67   0.095    -.1397732    1.741758
          7  |   .7503853   .4415144     1.70   0.089     -.114967    1.615738
          8  |   .7060934   .3956667     1.78   0.074     -.069399    1.481586
          9  |   .6660008   .3498308     1.90   0.057     -.019655    1.351657
         10  |    .628893   .3154035     1.99   0.046     .0107135    1.247072
         11  |   .5940541   .3062951     1.94   0.052    -.0062733    1.194381
------------------------------------------------------------------------------

. 
. 
. *** Import predicted values *** 
. use mars3.dta, clear
(Created by command margins; also see char list)

. 
. 
. *** Keep point estimates, upper bound, and lower bound ***
. keep _margin _ci_lb _ci_ub

. 
. 
. *** Generate counts of NGOs for the graph ***
. gen ngo_counts = 0

. replace ngo_counts = 5 in 2
(1 real change made)

. replace ngo_counts = 10 in 3
(1 real change made)

. replace ngo_counts = 15 in 4
(1 real change made)

. replace ngo_counts = 20 in 5
(1 real change made)

. replace ngo_counts = 25 in 6
(1 real change made)

. replace ngo_counts = 30 in 7
(1 real change made)

. replace ngo_counts = 35 in 8
(1 real change made)

. replace ngo_counts = 40 in 9
(1 real change made)

. replace ngo_counts = 45 in 10
(1 real change made)

. replace ngo_counts = 50 in 11
(1 real change made)

. 
. 
. *** Make a graph ***
. *
. * NOTE: I called the first tick a 0 even though you can't use the natural log on a 1. It's for graph continui
> ty. Feel free to change it 
. *
. graph twoway line _margin ngo_counts, clwidth(medium) clcolor(black) clpattern(solid) sort || line _ci_lb ngo
> _counts, clpattern(dash) clwidth(thin) clcolor(black) sort || line _ci_ub ngo_counts, clpattern(dash) clwidth
> (thin) clcolor(black) sort legend(off) xlabel(0 5 10 15 20 25 30 35 40 45 50) xtitle("Number of Women's Right
> s NGOs", size(medsmall)) ytitle("Predicted Count of SEA Misconduct", size(medsmall)) scheme(sj) title("Effect
>  of Women's Rights NGO Counts on Category 1 and SEA Misconduct", size(medium))

. graph export "Model_10.jpg", as(jpg) name("Graph") quality(90) replace
file /Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Model_10.jpg saved as JPG format

. 
. 
. 
. 
. 
end of do-file

. log close
      name:  <unnamed>
       log:  /Users/kellanrobinson/Desktop/Final Paper and Analysis for Submission 1/Final Log.log
  log type:  text
 closed on:   1 Dec 2023, 15:09:50
---------------------------------------------------------------------------------------------------------------
